Background:  The lack of biomarkers predictive of efficacy and safety is a critical gap in microbicide development. This was highlighted in a recent cellulose sulfate clinical trial, where treatment with the microbicide was associated with an increased risk of HIV acquisition. Measuring cytotoxicity in vitro does not suffice in terms of safety assessment of candidate compounds. The multilayered epithelium in the vagina and ectocervix provides the first line of defense against HIV and other pathogens and disruption of this barrier facilitates HIV transmission. We evaluated the effect of candidate microbicides on the architecture of polarized epithelial cell cultures and murine genital tract tissue; expression of immune mediators and the consequences for transmission of HIV.

Methods:  Human epithelial cells or reconstituted tissue were grown under polarized conditions in a dual chamber culture system. Following in vitro treatment of cultured cells or intravaginal application to the mouse genital tract (single or repeated) of cellulose sulfate, PRO 2000, or tenofovir, the effect on epithelial cell integrity and cellular junctions was evaluated by monitoring transepithelial electrical resistance (TER) and by confocal microscopy. Furthermore, changes in expression of structural proteins were detected at the RNA and protein levels. Finally, HIV addition to the apical chambers in vitro allowed assessment of virus migration across the epithelium to infect susceptible T cells cultured in the basolateral compartments.

Results:  Application of cellulose sulfate (100 µg/mL or 6% gel) led to a significant disruption of tight and adherens junctions in both models, inducing a rapid and sustained loss of TER across epithelial cells. The cellulose sulfate-induced disruption facilitated translocation of HIV across the epithelium as evidenced by confocal microscopy. In contrast, tenofovir had no deleterious effects. PRO 2000 caused a more modest loss in TER, but had no effect on adherens junction proteins and no increase in HIV translocation was observed. Repeated exposures to cellulose sulfate also triggered an inflammatory response characterized by the release of inflammatory cytokines and a loss of the protective protein secretory leukocyte protease inhibitor.

Conclusions:  Together, these findings provide a biological explanation that may have contributed to the failed cellulose sulfate clinical trial and indicate tenofovir safety. These data suggest the inclusion of the models presented in the preclinical evaluation of candidate microbicides.